You have purchase histories for 50,000 customers but no labels, and you want to find groups of similar customers. What kind of learning is this?
Options
- Unsupervised learning (clustering) (Correct option)
- Supervised regression
- Supervised classification
- Reinforcement learning
Answer
Unsupervised learning (clustering)
Why
Supervised learning needs labels (the right answers) to learn from. Here there are none, so you look for structure in the data itself.
Grouping similar items is clustering, the most common unsupervised task (k-means is a classic example).
Reinforcement learning is different: an agent learns by acting and receiving rewards.
Remember
No labels, find structure: unsupervised.